datos23 = pd.read_csv('https://raw.githubusercontent.com/Wilsonsr/metodos-estadisticos/main/inmuebles_bogota_res.csv', index_col=0) datos23

datos23.info() datos23.mnrobanos.unique()

datos23. mnrocuartos.unique()

datos23.describe(include="object")

datos23=datos23.drop_duplicates() datos23

datos23["LOgVentas"]=np.log10(datos23.mvalorventa) datos23["LOgVentas"]=np.log10(datos23.mvalorventa) datos23

import pandas as pd df2= pd.DataFrame(datos23) df2

import sklearn # Import label encoder from sklearn import preprocessing # label_encoder label_encoder = preprocessing.LabelEncoder() # df2["Codificada2"]=label_encoder.fit_transform(df2["mzona"]) df2

df2=pd.get_dummies(df2, columns=["mzona"]) df2

%%javascript IPython.OutputArea.auto_scroll_threshold = 9999;

df2.columns

p = sns.pairplot(df2, x_vars=["LOgmarea" , "mzona_Occidental" ,"mnrogarajes" ,"mnrobanos", "mnrocuartos" , "mzona_Centro" ,"mzona_Norte","mzona_Noroccidente","mzona_Sur"], y_vars="LOgVentas", size=7, aspect=0.7)

## encontrar el valor F critico import scipy.stats scipy.stats.f.ppf(q=1-0.05, dfn=18,dfd=1546)

modelo_multiple2=smf.ols("LOgVentas~ marea + mnrogarajes + mnrocuartos + mnrobanos ", data=df2).fit() modelo_multiple2.summary()

modelo_multiple3=smf.ols("LOgVentas ~ LOgmarea +mnrobanos+ mnrogarajes + mnrocuartos + mzona_Centro + mzona_Norte+mzona_Noroccidente+ mzona_Sur+ mzona_Occidental+ mzona_Chapinero", data=df2).fit() modelo_multiple3.summary()

modelo_multiple4=smf.ols("LOgVentas ~ LOgmarea+ mnrogarajes + mnrocuartos + mzona_Norte+ mzona_Sur+ mzona_Chapinero+ mzona_Occidental", data=df2).fit() modelo_multiple4.summary()

residuales=modelo_multiple4.resid residuales.mean()

residuales.sort_values(ascending=True).head(10)

df2.loc[719,]

df2.loc[[719,6],]

import matplotlib.pyplot as plt plt.subplots(figsize=(12, 6)) ax = plt.subplot(111) plt.scatter(x=residuales.index, y=residuales, alpha=0.5) plt.plot(np.repeat(0,residuales.index.max()), color='darkorange', linestyle='-') plt.title('Residuales') plt.show()

%matplotlib inline %config InlineBackend.figure_format ='retina' import seaborn as sns import matplotlib.pyplot as plt import statsmodels.stats.api as sms sns.set_style('darkgrid') sns.mpl.rcParams['figure.figsize'] = (15.0, 9.0) def homoscedasticity_test(model): ''' Function for testing the homoscedasticity of residuals in a linear regression model. It plots residuals and standardized residuals vs. fitted values and runs Breusch-Pagan and Goldfeld-Quandt tests. Args: * model - fitted OLS model from statsmodels ''' fitted_vals = model.predict() resids = model.resid resids_standardized = model.get_influence().resid_studentized_internal fig, ax = plt.subplots(1,2) sns.regplot(x=fitted_vals, y=resids, lowess=True, ax=ax[0], line_kws={'color': 'red'}) ax[0].set_title('Residuals vs Fitted', fontsize=16) ax[0].set(xlabel='Fitted Values', ylabel='Residuals') sns.regplot(x=fitted_vals, y=np.sqrt(np.abs(resids_standardized)), lowess=True, ax=ax[1], line_kws={'color': 'red'}) ax[1].set_title('Scale-Location', fontsize=16) ax[1].set(xlabel='Fitted Values', ylabel='sqrt(abs(Residuals))') bp_test = pd.DataFrame(sms.het_breuschpagan(resids, model.model.exog), columns=['value'], index=['Lagrange multiplier statistic', 'p-value', 'f-value', 'f p-value']) gq_test = pd.DataFrame(sms.het_goldfeldquandt(resids, model.model.exog)[:-1], columns=['value'], index=['F statistic', 'p-value']) print('\n Breusch-Pagan test ----') print(bp_test) print('\n Goldfeld-Quandt test ----') print(gq_test) print('\n Residuals plots ----')

homoscedasticity_test(modelo_multiple3)

p = sns.distplot(residuales,kde=True) p = plt.title('Normality of error terms/residuals')

import statsmodels.api as sm # Gráfico Q-Q # ============================================================================== fig, ax = plt.subplots(figsize=(7,4)) sm.qqplot( residuales, fit = True, line = 'q', alpha = 0.4, lw = 2, ax = ax ) ax.set_title('Gráfico Q-Q Residuales', fontsize = 10, fontweight = "bold") ax.tick_params(labelsize = 7)

from scipy import stats # Shapiro-Wilk test # ============================================================================== shapiro_test = stats.shapiro(residuales) shapiro_test

KM= stats.ks_1samp(residuales, stats.norm.cdf ) KM

import statsmodels.tsa.api as smt acf = smt.graphics.plot_acf(residuales, lags=40 , alpha=0.05) acf.show()

df2

plt.figure(figsize=(25,25)) p=sns.heatmap(df2.corr(), annot=True,square=True)

df2.describe

VIf_area=(1/1-0.78) VIf_garajes=(1/1-0.61) VIf_garajes ,VIf_area

from statsmodels.stats.outliers_influence import variance_inflation_factor X= df2[[ 'mnrogarajes', 'marea', 'mzona_Norte','mnrocuartos','mzona_Sur', 'mzona_Centro','mzona_Chapinero',"mzona_Occidental"]] # VIF dataframe vif_data = pd.DataFrame() vif_data ["feature"] = X.columns # calculating VIF for each feature vif_data["VIF"] = [variance_inflation_factor(X.values, i) for i in range(len(X.columns))] print(vif_data)

Creamos una tabla con los valores de las características para predicir con el módelo selccionado modelo_multiple4.

LOgVentas ~ LOgmarea + mnrogarajes + mnrocuartos